Historical inputs from fossil fuel and biomass

0 downloads 0 Views 601KB Size Report
1950s, while molecular and elemental evidences pointed to oil .... replicate analyses of NIST SRM 1649 and 1649a (urban dust) was w5%. 3. Results and ...

Environmental Pollution 159 (2011) 983e990

Contents lists available at ScienceDirect

Environmental Pollution journal homepage: www.elsevier.com/locate/envpol

Combustion-derived substances in deep basins of Puget Sound: Historical inputs from fossil fuel and biomass combustion Li-Jung Kuo a, *,1, Patrick Louchouarn b, c, Bruce E. Herbert a, Jill M. Brandenberger d, Terry L. Wade e, Eric Crecelius d a

Department of Geology & Geophysics, Texas A&M University, College Station, TX 77843, USA Department of Marine Science, Texas A&M University at Galveston, Galveston, TX 77551, USA Department of Oceanography, Texas A&M University, College Station, TX 77843, USA d Pacific Northwest National Laboratory, Marine Science Laboratory, Sequim, WA 98382, USA e Geochemical & Environmental Research Group, Texas A&M University, College Station, TX 78433, USA b c

Temporal trend of GBC was directly linked to human activities, while the input of char-BC in Puget Sound was more likely driven by regional climate oscillations.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 14 September 2010 Received in revised form 30 November 2010 Accepted 3 December 2010

Reconstructions of 250 years historical inputs of two distinct types of black carbon (soot/graphitic black carbon (GBC) and char-BC) were conducted on sediment cores from two basins of the Puget Sound, WA. Signatures of polycyclic aromatic hydrocarbons (PAHs) were also used to support the historical reconstructions of BC to this system. Down-core maxima in GBC and combustion-derived PAHs occurred in the 1940s in the cores from the Puget Sound Main Basin, whereas in Hood Canal such peak was observed in the 1970s, showing basin-specific differences in inputs of combustion byproducts. This system showed relatively higher inputs from softwood combustion than the northeastern U.S. The historical variations in char-BC concentrations were consistent with shifts in climate indices, suggesting an influence of climate oscillations on wildfire events. Environmental loading of combustion byproducts thus appears as a complex function of urbanization, fuel usage, combustion technology, environmental policies, and climate conditions. Ó 2010 Elsevier Ltd. All rights reserved.

Keywords: Black carbon Char Levoglucosan Climate oscillations Pacific Northwest

1. Introduction Black carbon (BC) is a generic term for the carbonaceous residue from incomplete combustion of organic matter and has been found widely distributed in many surficial reservoirs, such as soils, sediments, water bodies, and the atmosphere (Goldberg, 1985). BC can be described as a continuum of physically and chemically heterogeneous materials ranging from slightly charred materials to highly refractory soot particles with increasing combustion temperature (Masiello, 2004; Hammes et al., 2007). In brief, the BC continuum comprises two major categories, char/charcoal BC (collectively referred to as char-BC) and soot/graphitic BC (collectively referred to as GBC), with biomass burning and fossil fuel combustion as the major sources (Schmidt and Noack, 2000). Because of its long

* Corresponding author. E-mail address: [email protected] (L.-J. Kuo). 1 Current address: Pacific Northwest National Laboratory, Marine Science Laboratory, Sequim, WA 98382, USA 0269-7491/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.envpol.2010.12.012

environmental lifetime and combustion origin (Masiello and Druffel, 1998), the presence of BC in environmental archives (sediments, soils, ice cores) is an ideal proxy for reconstructing the history of industrialization and shifts in combustion technology/ fuel types (Elmquist et al., 2007; Louchouarn et al., 2007) as well as paleofire history due to climate oscillations (Wolbach et al., 1988; Bird and Cali, 1998). Environmental loadings of BC and co-generated toxic pollutants such as polycyclic aromatic hydrocarbons (PAHs) have been tied to anthropogenic activities since the Industrial Revolution due to the associations between combustion practices and the rapid increases in human populations and energy usage (Elmquist et al., 2007; Louchouarn et al., 2007). To understand the linkages between the onset and development of industrialization and inputs of BC and PAHs to the environment, as well as to evaluate the effectiveness of environmental regulations in the past century and assess current environmental status, many studies have used sediment or ice cores to reconstruct the temporal trends of these combustion byproducts (Gustafsson et al., 1997; Buckley et al., 2004; Wakeham et al., 2004; Muri et al., 2006; Elmquist et al., 2007; Louchouarn

984

L.-J. Kuo et al. / Environmental Pollution 159 (2011) 983e990

et al., 2007; McConnell et al., 2007). These studies have successfully confirmed the significant increase of BC emissions in the earlyto-mid 20th century and its later decrease due to changes in fuel usages and improvement of combustion efficiency as well as emission reductions. However, discrepancies in the historical BC records across studies have also been revealed. Cores from different regions show that BC maxima occurred at different times and were connected to different sources. For example, a soot maximum in a rural Swedish lake was observed in 1920 and linked to the predominance of wood combustion at the time (Elmquist et al., 2007). In comparison, the same type of black carbon in a sediment core from Central Park Lake in New York City peaked around 1950s, while molecular and elemental evidences pointed to oil combustion and incineration of municipal solid waste as predominant sources (Louchouarn et al., 2007). The heterogeneity in the historical and geographical distribution of BC around the world might be the result of a combination of regional differences in fuel use, differences in combustion technologies, and the relatively short lifetime of BC in the atmosphere (Ogren and Charlson, 1983; Bond et al., 2004; Louchouarn et al., 2007). While many temporal BC reconstructions have been conducted in the northeastern U.S. and Europe (Gustafsson et al., 1997; Buckley et al., 2004; Muri et al., 2006; Elmquist et al., 2007; Louchouarn et al., 2007), studies in the Pacific Northwest (PNW) are scarce (Wakeham et al., 2004). This region is of particular interest because of its late settlement and thus its different fuel usage history, as well as the distinct climate system compared to northeastern U.S. One of the few studies for the Seattle region, WA (Wakeham et al., 2004) showed that combustion-derived PAHs and GBC peaked synchronously in the early-to-mid 1950s in a core from Lake Washington, suggesting a similar source for both. In the present study, we conducted a more comprehensive investigation on historical BC deposition records from three dated sediment cores in the Main Basin of Puget Sound and in the Hood Canal, WA. The metropolitan area of Seattle/Tacoma, a major urban coastal system in the PNW, is located on the eastern shore of the Main Basin. The population of the three coastal counties (King, Pierce, and Kitsap) around the Main Basin has grown rapidly over the last 50 years compared to the sparsely populated counties surrounding Hood Canal (Office of Financial Management Washington State, 2010). We conducted the GBC analysis to reconstruct decadalscale inputs of the high temperature-derived, highly condensed BC to these cores. In addition, we measured levoglucosan, a biomarker derived mostly from biomass combustion (Kuo et al., 2008), to reconstruct the historical inputs of low temperature char-BC, a BC type hard to isolate using most BC methods. Finally, concentrations and diagnostic ratios of PAHs were also used to support the historical reconstructions of BC to this system. This multi-core, multi-proxy study enables us to identify the spatial and temporal distributions of diverse combustion-derived materials across this region and gain further insight into (i) the effects of urbanization/ industrialization on BC inputs to local coastal basins (Main Basin vs. Hood Canal), (ii) the potential differences in regional distribution patterns of BC (PNW vs. northeastern U.S.), and (iii) the governing factors for the production of different BC classes (GBC and char-BC).

2. Materials and methods 2.1. Site description and sediment collection The Main Basin of Puget Sound (250 m depth) is a 70 km-long urban fjord-like estuary bounded on the north by a sill at Admiralty Inlet (66 m) and on the south by a sill at the Tacoma Narrows (44 m). Separated from Puget Sound Main Basin by the Kitsap Peninsula, Hood Canal (50e150 m depth) is also a glacially carved fjord. It opens from Admiralty Inlet at its northern limit and stretches about 110 km to its southern end (The Great Bend).

Sediment cores were collected in deep waters from two sites in the Main Basin (PS-1, near Tacoma; PS-4, near Seattle) and one in Hood Canal (HC-3; see Figure S1) in 2005 using a stainless steel, open barrel gravity corer (Kasten corer) as described elsewhere (Brandenberger et al., 2008). Coring was conducted aboard the University of Washington research vessel, RV Barnes. After collection, sediment in contact with the core barrel was removed prior to subsection. Sediment cores from PS-1 and PS-4 were sectioned at 2 cm intervals from 0 to 60 cm and 0e50 cm, respectively, then at 5 cm intervals down to the core catcher (220e260 cm). The HC-3 core was sampled every 2 cm throughout the entire core (215 cm). The subsectioned sediments were immediately collected into precleaned glass jars, and stored in coolers until transport back to the PNNL Marine Science Laboratory in Sequim, WA. Each core segment was homogenized and freeze-dried for further analyses. 2.2. Sediment dating The geochronology of sediment cores was established using steady-state 210Pb dating techniques described in our earlier works (Bloom and Crecelius, 1987; Brandenberger et al., 2008). Sedimentation rate (cm yr1) of each core was estimated by measuring the alpha activity of the granddaughter product 210Po and timecorrected to 210Pb assuming both 210Po and 210Pb remain in secular equilibrium in the post-depositional environment. The chronology of the Main Basin sites has been confirmed through independent approaches including stable inorganic tracers, radiometric dating, and two decades of repeated chronological sampling at the same stations, producing sedimentation rates that are consistent with previously published values for the area (Bates et al., 1984; Lefkovitz et al., 1997; Brandenberger et al., 2008). The consistency in sedimentation rate derived from the overlap of stable metal profiles from repeated corings at the same sites (Brandenberger et al., 2008) was reproduced using PAH profiles (see discussion below), confirming the numerous crosschecks of geochronologies in these sediment cores (Brandenberger et al., 2008, Brandenberger et al., submitted for publication). 2.3. GBC analysis The high temperature-derived BC in the sediment was quantified with the GBC method developed by Gélinas et al. (2001) with modifications (Louchouarn et al., 2007) as described in the supplementary information (SI). The GBC method is a chemo-thermal oxidation (CTO) procedure using rigorous stepwise demineralization and hydrolysis treatments before thermal oxidation. This method can minimize the potential analytical artifacts generated sometimes by the conventional CTO method (CTO-375), which results in the charring of labile organic matter during the treatment of carbon-rich samples (Gélinas et al., 2001). A potential drawback of GBC method is that mass loss may happen during sample manipulation (Elmquist et al., 2004). However, Louchouarn et al. (2007) have shown comparable black carbon values from GBC and CTO-375 methods in environmental samples with wide ranging values of OC and BC. Similar to most CTO methods, the analytical window of the GBC method only quantifies the highly condensed BC and excludes char-BC derived from low temperature combustion (Hammes et al., 2007; Kuo et al., 2008). As a quality control approach, two NIST standard reference materials (SRM 1941b and SRM 1944) were analyzed regularly along with the sediment samples. The average GBC values for these SRMs are 0.39  0.09 and 0.78  0.04%, respectively, which are within the range of reported values from GBC and CTO-375 methods (Hammes et al., 2007; Louchouarn et al., 2007). The relative standard deviation (RSD) of the duplicate analysis of selected sediments was